The nonreceptor protein-tyrosine kinase c-Fes is involved in fibroblast growth factor-2-induced chemotaxis of murine brain capillary endothelial cells.

Fibroblast growth factor-2 (FGF-2)-induced migration of endothelial cells is involved in angiogenesis in vivo. However, signal transduction pathways leading to FGF-2-induced chemotaxis of endothelial cells are largely unknown. Previous studies have shown that the cytoplasmic protein-tyrosine kinase c-Fes is expressed in vascular endothelial cells and may influence angiogenesis in vivo. To investigate the contribution of c-Fes to FGF-2 signaling, we expressed wild-type or kinase-inactive human c-Fes in the murine brain capillary endothelial cell line, IBE (Immortomouse brain endothelial cells). Wild-type c-Fes was tyrosine-phosphorylated upon FGF-2-stimulation in transfected cells, whereas kinase-inactive c-Fes was not. Overexpression of wild-type c-Fes promoted FGF-2-independent tube formation of IBE cells. Tube formation was not observed with endothelial cells expressing kinase-inactive c-Fes, indicating a requirement for c-Fes kinase activity in this biological response. Expression of kinase-defective c-Fes suppressed endothelial cell migration following FGF-2 treatment, suggesting that activation of endogenous c-Fes may be required for the chemotactic response. Expression of either wild-type c-Fes or the kinase-inactive mutant did not affect the tyrosine phosphorylation FRS2, Shc, or phospholipase C-gamma, nor did it influence the kinetics of mitogen-activated protein kinase activation. These results implicate c-Fes in FGF-2-induced chemotaxis of endothelial cells through signaling pathways not linked to mitogenesis.

Angiogenesis is involved in many physiological and pathological processes, such as ovulation, embryogenesis, malignant tumor growth, retinopathies, and rheumatoid arthritis (1)(2)(3). In angiogenesis, activated endothelial cells produce proteases that digest the basement membrane of blood vessels, allowing them to migrate into interstitial tissue, proliferate, and finally form lumen-containing, tube-like structures (1,4). These biological responses are tightly regulated by fibroblast growth factors (FGFs) 1 and vascular endothelial growth factors, which mediate their effects by binding to specific receptor tyrosine kinases (5)(6)(7). These receptor tyrosine kinases are linked to a number of downstream effector molecules including components of the ubiquitous Ras/mitogen-activated protein kinase (Ras/MAPK) pathway and phospholipase C-␥.
The FGF receptor (FGFR) subfamily of receptor tyrosine kinases consists of four structurally related members, designated FGFR-1 through FGFR-4 (8,9). All members of the family are composed of an extracellular domain with two or three immunoglobulin-like loops, a single transmembrane domain, and an intracellular kinase domain, which is divided by an insert region. Seven tyrosine residues in the cytoplasmic region of FGFR-1 have been identified as autophosphorylation sites (10). Of these, Tyr-766 in the C-terminal tail of FGFR-1 binds to the Src homology 2 domain of PLC-␥, leading to its tyrosine phosphorylation and activation (11,12). Substitution of Tyr-766 with Phe resulted in slower internalization of FGFR-1, suggesting a function in receptor turnover (13).
FGFR-1 is also linked to several small G-protein/kinase cascades. Activation of the Ras pathway has been implicated in both proliferation and tube formation of cultured endothelial cells (14). The adaptor proteins Shc and FRS2 are tyrosinephosphorylated by FGF-2 treatment, leading to recruitment of the Grb2⅐Sos guanine nucleotide exchange complex for Ras (14 -16). Activated Ras binds to Raf-1, leading to activation of the extracellular signal regulated kinases Erk-1 and Erk-2 through the intermediate kinase mitogen-activated protein kinase/Erk kinase-1. The farnesyl transferase inhibitor, manumycin, and the prenylated protein methyltransferase inhibitor, FTS, both attenuate tube formation, presumably by interfering with Ras function (14). The mitogen-activated protein kinase/ Erk kinase inhibitor PD98059 suppressed proliferation of endothelial cells without affecting tube formation (14), suggesting that the differentiation signal diverges above the level of mitogen-activated protein kinase/Erk kinase activation. In addition to Shc, FRS2, and Grb2⅐Sos, the adaptor protein Crk also binds to the activated FGFR-1 through phosphotyrosine 463 in the juxtamembrane region. Crk in turn recruits C3G, a guanine nucleotide exchanger factor for Rap1 (17). Substitution of Tyr-463 with Phe disrupts FGF-2-induced DNA synthesis (17), implicating the Crk⅐C3G interaction in the mitogenic response.
The c-Fes protein-tyrosine kinase is structurally distinct from c-Src, c-Abl, and other nonreceptor tyrosine kinases (reviewed in Ref. 18). c-Fes is expressed predominantly in hematopoietic cells, vascular endothelial cells, and some epithelial and neuronal cells (19). Several lines of evidence suggest that c-Fes may play a direct role in myeloid differentiation. Expression of c-Fes in the myeloid leukemia cell line K-562, an immature blast cell line devoid of c-Fes expression, results in growth suppression and terminal differentiation (20,21). This result suggests that c-Fes expression is sufficient for differentiation in some cellular contexts. On the other hand, suppression of c-Fes expression with antisense oligonucleotides interferes with myeloid differentiation in HL-60 promyelocytes, and in some cases, it results in apoptosis instead (22,23). Thus, c-Fes may be required for differentiation to occur. c-Fes may regulate differentiation in extrahematopoietic sites as well. Expression of an activated form of c-Fes bearing the N-terminal myristylation signal sequence from v-Src produced a marked increase in vascularity and hemangioma formation in transgenic mice (24). This result indicates that c-Fes is involved in vascular development in vivo. However, the precise roles of c-Fes in the regulation of cellular responses by endothelial cells are unknown.
In this report, we examined the effect of c-Fes expression on FGF-2-induced proliferation, chemotaxis, and tube formation of IBE cells, a murine brain capillary endothelial cell line (25). Transfected c-Fes was autophosphorylated upon FGF-2 stimulation, suggesting that FGF-2 may regulate c-Fes kinase activity in vivo. Expression of a kinase-inactive c-Fes mutant interfered with FGF-2-induced chemotaxis of IBE cells, suggesting that endogenous c-Fes activation may be important for this response. In addition, tube formation by IBE cells was stimulated by overexpression of wild-type c-Fes in the absence of FGF-2 treatment. These results indicate that c-Fes is both necessary and sufficient for FGF-2-induced cellular responses by endothelial cells.
Transfection of FLAG-tagged c-Fes Constructs into IBE Cells-Liposomes, denoted TFL-3, -5, -6, and -8, were kindly provided by Dr. Hiroshi Kikuchi (Daiichi Pharmaceuticals, Tokyo, Japan). Empty pcDNA3 (Invitrogen, San Diego, CA) or expression plasmids containing C-terminally FLAG-tagged wild-type c-Fes or kinase-inactive (K590E) c-Fes were used (26). Dried liposomes were resuspended in sterile Milli-Q water containing purified plasmid DNA (1 g of DNA/10 M cationic lipid) and left for 15 min at room temperature. IBE cells (2 ϫ 10 6 cells/10-cm dish) were incubated with Ham's F-12 medium containing 1.5 g/ml plasmid DNA for 7 h, and then the medium was changed to a fresh growth medium. On the third day, 0.4 mg/ml G418 (Life Technologies, Inc.) was added, and the culture was continued for 10 days. After selection, G418-resistant clones were expanded for analysis of expression of the c-Fes-FLAG constructs by immunoblotting.
Cell Proliferation Assay-The cell proliferation assay was performed as described previously, with some modifications (27). Briefly, transfected IBE cells were plated in human plasma fibronectin-coated 24well culture plates at a density of 1.5 ϫ 10 4 cells/cm 2 (3 ϫ 10 4 cells/well) in Ham's F-12 medium containing 10% fetal bovine serum and cultured at 33°C. The next day, the medium was changed to Ham's F-12 medium containing 0.25% fatty acid-free bovine serum albumin (BSA) with or without FGF-2, and the culture was continued for 3 days. Cells were detached with trypsin, and the cells were counted using a hemocytometer. The number of cells in the untreated samples was set to 100%.
Chemotaxis Assay-Chemotaxis assay was performed using Transwell membrane filters (Corning Costar Japan, Tokyo, Japan). Membranes (polycarbonate, 8 m pores) were coated with 10 g/ml fibronectin. Ham's F-12 medium containing 0.25% BSA with or without the indicated concentrations of FGF-2 was added into lower wells of 24-well plates. Transfected IBE cells were harvested with trypsin, incubated with soybean trypsin inhibitor (Sigma), and washed. Cells were resuspended in Ham's F-12 medium containing 0.25% BSA at a density of 3 ϫ 10 5 cells/ml, plated on the upper wells (100 l/well), and incubated for 4 h at 33°C. Cells were washed, fixed in 100% methanol for 10 min at room temperature, washed again, and stained with Giemsa solution. Cells on the upper surface of the membranes were removed with cotton swabs, and the cells attached to the lower surface of the membranes were counted microscopically. The number of cells counted in untreated samples was set to 100%.
Zymographic Assay for the Determination of Plasminogen Activator (PA) Activity-Zymographic assay for the determination of secreted PA activity was described previously (25). In brief, IBE cells were plated in fibronectin-coated 24-well plates and cultured in growth medium for 24 h at 33°C. Once confluent, the cells were washed with Ham's F-12 medium three times and cultured in Ham's F-12 medium containing 0.25% BSA for 24 h. The medium was then replaced with fresh Ham's F-12 medium containing 0.25% BSA with or without indicated concentrations of FGF-2 and cultured for 24 h. The culture medium was collected and centrifuged to remove cell debris, and aliquots of the medium were electrophoresed in SDS-polyacrylamide gels (10% polyacrylamide) under nonreducing conditions. Gels were washed extensively with phosphate-buffered saline containing 2.5% Triton X-100 and put horizontally on top of a 1.1% agarose gel containing 1.6% nonfat milk (as a source of casein) and 0.12 units/ml plasminogen (Calbiochem, La Jolla, CA) and incubated at 37°C for 18 h. PA activity was visualized as translucent areas in the white agarose gel. The gels were photographed, and quantitation was performed using the NIH Image program, version 1.52. The intensity of the band in lanes of untreated cells was set to 100%.
Tube Formation Assay-Tube formation assay was performed as described elsewhere (25). Briefly, IBE cells were cultured between two layers of type I collagen gels without serum or growth supplements in the presence or absence of 10 ng/ml FGF-2 at 33°C for 18 h. Cultures were photographed under the phase-contrast microscope.
Immunoprecipitation and Immunoblotting-IBE cells grown to confluence on fibronectin-and gelatin-coated dishes were growth supplement-starved with Ham's F12 containing 20 units/ml aprotinin and 0.25% BSA overnight, followed by treatment in the presence or absence of 100 ng/ml FGF-2 for 8 min at 33°C. Cells were washed once with Tris-buffered saline, pH 7.5, containing 100 M orthovanadate on ice and lysed in lysis buffer (50 mM Tris-HCl, pH 7.5, containing 0.15 M NaCl, 1% Triton X-100, 10 mM NaF, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, 0.02% SDS, 100 M orthovanadate, and 100 units/ml aprotinin). Clarified cell lysates were incubated with either anti-FLAG M2 antibody, anti-PLC-␥ antibody, or anti-Shc antibody followed by the adsorption to protein A-agarose beads. Alternatively, p13 suc1 -agarose beads (Upstate Biotechnologies, Inc.) were used to precipitate FRS-2. After washing, proteins were eluted from beads by heating in SDSsample buffer and then separated by SDS-polyacrylamide gel electrophoresis. Proteins were electrophoretically transferred to polyvinylidene difluoride membranes (Millipore, Bedford, MA) and probed with the indicated antibodies. Antibody incubation was followed by incubation with horseradish peroxidase-conjugated anti-rabbit IgG or antimouse IgG, and detection was through enhanced chemiluminescence reaction (ECL, Amersham Pharmacia Biotech). Membranes were stripped of antibodies by soaking in 62.5 mM Tris-HCl, pH 6.8, containing 2% SDS and 0.7% 2-mercaptoethanol at 50°C for 30 min. For some blots, total cell lysates were prepared by heating the cells in SDS sample buffer, and 50 g of the resulting proteins extracts were electrophoresed and analyzed.

FGF-2 Induces
Tyrosine Phosphorylation of c-Fes-The model system for all studies was the IBE capillary endothelial cell line, which was established from the brains of transgenic mice expressing a temperature-sensitive mutant of SV40 large T antigen. At the permissive temperature (33°C), IBE cells proliferate, migrate, secret urokinase-type plasminogen activator (u-PA),

c-Fes in Migration of Endothelial Cells
and form tube-like structures in response to FGF-2 treatment. IBE cells express mainly FGFR-1 with small amounts of FGFR-2 and -4, and the signals for FGF-2-induced biological responses require the intracellular region of the FGFR-1 (25).
Recent studies have established marked c-Fes expression in vascular endothelial cells (19). To examine the expression of endogenous c-Fes in IBE cells, total cell lysates were prepared from IBE cells and probed for expression of endogenous c-Fes by immunoblotting. As a positive control, equivalent protein extracts were prepared from human umbilical cord vein endothelial cells, which have been shown previously to express high levels of c-Fes (24). As shown in Fig. 1A, endogenous c-Fes expression was readily detectable in both cells. To examine the role of c-Fes in FGF-2-induced cellular responses of IBE cells, FLAG-tagged wild-type or kinase-inactive c-Fes expression plasmids were introduced into IBE cells, and cell lines were obtained by G418 selection. To verify expression of c-Fes in the transfected cells, anti-FLAG immunoprecipitates were prepared from soluble cell lysates and probed with the FLAG antibody. Fig. 1B shows the result with two clones expressing wild-type Fes (clones 6-1 and 6-8) and clones expressing kinase-inactive Fes (clones 5-8 and 5-15) that were selected for further examination. The expression of wild-type Fes was consistently lower than that of the kinase-defective mutant in all of the clones examined, although the reason for this observation is not apparent.
To determine whether c-Fes tyrosine kinase activity is affected by FGF-2, the transfected IBE cells were treated with FGF-2 and c-Fes proteins were immunoprecipitated and analyzed for phosphotyrosine content by anti-phosphotyrosine immunoblotting. Fig. 1C shows that wild-type c-Fes was tyrosinephosphorylated upon FGF-2 stimulation. In contrast, FGF-2 stimulation was unable to induce tyrosine phosphorylation of the kinase-defective mutant of c-Fes. These results suggest that c-Fes is activated by FGF-2 treatment and may function to transmit signals for growth or differentiation. (28,29). These results raised the possibility that Src family kinases may act as intermediates between the FGF receptor and activation of c-Fes downstream. To test this possibility in a defined system, we co-expressed the kinase-defective form of c-Fes with c-Src in Sf-9 insect cells and assessed c-Src-induced Fes phosphorylation on anti-phosphotyrosine immunoblots. As shown in Fig. 2, immunoprecipitates from cells expressing the kinase-inactive c-Fes mutant alone showed no detectable phosphotyrosine. However, co-expression with wild-type c-Src and the related kinase Fyn resulted in potent tyrosine phosphorylation of kinase-inactive c-Fes, comparable to the level attained by autophosphorylated active c-Fes. These results raise the possibility that c-Fes activation by FGF-2 treatment may occur by both direct and indirect mechanisms.

c-Fes Specifically Transduces Signals for FGF-2-induced Chemotaxis of Endothelial Cells-Although expression of c-Fes
in transgenic animals leads to marked stimulation of angiogenesis, the impact of c-Fes activation at the level of endothelial cells has not been determined. c-Fes activation may affect cell proliferation, PA secretion, cellular migration and tube formation, which together make up the major components of the angiogenic response. We first examined the effect of FGF-2 on proliferation in IBE cells expressing wild-type and kinase-defective c-Fes. Fig. 3 shows that expression of wild-type c-Fes had little effect on the proliferation of IBE cells, whereas the kinase-defective mutant produced a modest but reproducible suppression of growth. We next investigated whether c-Fes influences FGF-2-mediated u-PA secretion by IBE cells. Cells were treated with FGF-2 at both 5 and 50 ng/ml, and u-PA activity in the culture supernatant was examined using a zymographic assay. As shown in Fig. 4, FGF-2-stimulated u-PA secretion was unaffected by the presence of either form c-Fes. This result suggests that signals for u-PA secretion in response to FGF-2 treatment do not involve c-Fes.
We next assessed the role of c-Fes in the chemotaxic response of endothelial cells to FGF-2. As shown in Fig. 5, FGF-2 stimulated marked chemotaxis of mock-transfected IBE cells in a dosedependent manner, consistent with previous results (25). Expression of wild-type c-Fes did not affect the chemotaxic response toward FGF-2. In contrast, expression of the kinase-inactive mutant of c-Fes completely inhibited the chemotaxic response of IBE cells to FGF-2. Previous studies have shown that this kinaseinactive mutant of c-Fes suppresses wild-type c-Fes autophosphorylation in vitro (30), suggesting that it may be capable of suppressing endogenous c-Fes activation by FGF-2 in IBE cells. These results implicate c-Fes in FGF-2-induced chemotaxis.
In a final series of experiments, FGF-2-induced tube forma-

FIG. 3. Overexpression of WT or kinase-inactive (K590E) c-Fes has little effect on FGF-2-induced proliferation of IBE cells.
Mock, K590E, or WT c-Fes-transfected IBE cells were plated in fibronectin-coated wells of 24-well plates in Ham's F-12 medium supplemented with 10% fetal bovine serum. The next day, the medium was replaced with Ham's F-12 medium containing 0.25% BSA with or without FGF-2. Three days later, cells were detached with trypsin, and cell number was counted using a hemocytometer. Data are shown as the means ϩ S.D. for triplicate wells. Comparable data were obtained from two independent experiments. Open bars, no FGF-2; hatched bars, 5 ng/ml FGF-2; solid bars, 50 ng/ml FGF-2.

FIG. 4. Overexpression of WT or kinase-inactive (K590E) c-Fes
has no effect on FGF-2-induced u-PA secretion. Mock, K590E or WT c-Fes-transfected cells were plated in fibronectin-coated wells of 24-well plates in growth medium. The next day, the medium was replaced with Ham's F-12 medium containing 0.25% BSA, and cells were cultured for 24 h. The medium was changed again to Ham's F-12 medium containing 0.25% BSA with or without FGF-2, and the culture was continued for another 24 h. Culture supernatants were collected and centrifuged to remove cell debris, and aliquots of the supernatants were electrophoresed under nonreducing conditions. After washing with phosphate-buffered saline containing 2.5% Triton X-100 to remove SDS, the polyacrylamide gels were overlaid on casein-and plasminogen-containing agarose gels and incubated for 16 h at 37°C. PA activity was visualized as translucent areas. Gels were placed onto black paper and photographed. Quantitation of each band was performed using the NIH Image program, version 1.52, and the relative intensity values are shown below the photograph.

c-Fes in Migration of Endothelial Cells
tion was examined using the transfected cell lines. As shown in Fig. 6, expression of wild-type c-Fes alone was sufficient to induce tube formation in the absence of FGF-2. Interestingly, expression of the kinase-inactive form of c-Fes did not interfere with FGF-2-induced tube formation as observed above for chemotaxis. This result suggests that redundant pathways exist for tube formation stimulated by FGF-2.

Effect of c-Fes Expression on FGF-2-induced Tyrosine Phosphorylation of PLC-␥ and Activation of Ras/MAPK Pathway-
Previous studies have shown that c-Fes activates a variety of signal transduction pathways, including the Ras/MAPK pathway (18). FGF-2 activates FGFR-1, leading to activation of Ras/MAPK pathway as well, by recruitment and phosphorylation of the adaptor proteins FRS2 and Shc (see Introduction).
To determine whether wild-type or kinase-inactive c-Fes influences these FGFR-mediated signaling events, FRS2 and Shc were precipitated from the control and c-Fes-expressing IBE cell lines and analyzed for tyrosine phosphorylation by immunoblotting. As shown in Fig. 7, FGF-2 induced potent phosphorylation of both FRS-2 and Shc, and these phosphorylation events were unaffected by the presence of either form of c-Fes. We also investigated the kinetics of MAPK activation in each of the cell lines and found that both immediate (8 min) and sustained (120 min) activation of MAPK were not altered by either form of c-Fes (Fig. 7C). We also examined the effects of c-Fes expression on the FGF-induced tyrosine phosphorylation of PLC-␥, another major substrate for the FGF receptor. Fig.  7D shows that tyrosine phosphorylation of PLC-␥ was equally stimulated by FGF-2 in control cells and cells expressing the wild-type and kinase-inactive forms of c-Fes. In addition, both PI3K activity and Stat3 tyrosine phosphorylation were unaffected by the presence of c-Fes in these cells (data not shown). These data indicate that the effects of c-Fes on IBE cell chemotaxis and tube formation involve a mechanism independent of the mitogenic pathways known to be activated by c-Fes in other systems (see under "Discussion").

DISCUSSION
In this report, we show for the first time that the nonreceptor protein-tyrosine kinase encoded by the c-fes proto-oncogene plays a role in FGF-2-induced angiogenic responses in cultured vascular endothelial cells. Overexpression of wild-type c-Fes in IBE capillary endothelial cells was sufficient to induce tube formation in the absence of FGF-2. On the other hand, expression of a kinase-defective mutant of c-Fes completely blocked FGF-2-induced chemotaxis, suggesting that activation of endogenous c-Fes may be required for this FGF-2 response. In contrast, neither wild-type nor kinase-defective c-Fes markedly influenced the proliferative response of IBE cells or expression of u-PA. These results suggest that c-Fes may be responsible for generating signals for cellular movement and differentiation and are consistent with previous data showing that expression of a membrane-targeted form of c-Fes results in hypervascularization in transgenic mice (24).
At the molecular level, we observed that FGF-2 was able to stimulate tyrosine phosphorylation of c-Fes in the transfected IBE cells. One mechanism to explain these data involves recruitment of c-Fes to the activated receptor and stimulation of c-Fes autophosphorylation. Other studies have shown that autophosphorylation of c-Fes on Tyr-713 within its kinase domain activation loop is essential for full kinase activity and substrate phosphorylation (31). Interestingly, we also observed that Src kinases strongly phosphorylate a kinase-defective form of c-Fes in Sf-9 cells (Fig. 2). This result suggests that activation of c-Fes may occur downstream of FGF-2-induced Src family kinase activation. Other work from our laboratory has shown that the Src family kinase inhibitor PP1, as well as overexpression of kinase-inactive c-Src, inhibited chemotaxis toward FGF-2 by IBE cells. 2 Together with the strong phosphorylation 2 T. Shono, H. Kanetake, and S. Kanda, submitted for publication.  We also investigated whether signaling pathways known to be activated by FGF-2 are affected by overexpression of wildtype or kinase-inactive c-Fes in IBE cells. One major substrate for FGFR-1 is PLC-␥, which binds through its Src homology 2 domain to the activated receptor via Tyr(P)-766. Although substitution of FGFR-1 Tyr-766 with Phe blocks recruitment and activation of PLC-␥, this receptor mutant still retains the ability to transduce signals for chemotaxis (32). In this study, overexpression of both wild-type and kinase inactive c-Fes had no effect on tyrosine phosphorylation of PLC-␥ in response to FGF-2 treatment. This result also supports the idea that PLC-␥ is not involved in FGF-2-induced cellular migration. Other studies have shown that both wortmannin and LY294002, which are potent inhibitors of PI3K, inhibit chemotaxis of porcine aortic endothelial cells (32). However, we observed that LY294002 inhibited PDGF-BB-induced chemotaxis but not FGF-2-induced chemotaxis in both control and c-Fes-expressing IBE cells (data not shown). Although c-Fes has been shown to activate PI3K in other systems (33), this result suggests that activation of PI3K is not required for FGF-2-induced chemotaxis of IBE cells. We also looked at signaling molecules that have been implicated in c-Fes signaling in fibroblast transformation systems, including the Ras/MAPK pathway and Stat3 activation (34,35). These pathways were unaffected under conditions in which wild-type or kinase-defective c-Fes produced biological effects on FGF-2-dependent responses in IBE cells. Taken together, these data suggest that c-Fes signals for angiogenesis in vascular endothelial cells are distinct from the mitogenic pathways known to be stimulated in fibroblasts transformed by activated forms of c-Fes (e.g. PI3K pathway, Stat3 activation, and small G protein/MAPK pathways). Similar findings have been made in macrophage cell lines overexpressing c-Fes. In these studies, phosphorylation of p130 Cas and other proteins related to cellular adhesion and cell-cell contact have been reported in the absence of phosphorylation of signaling molecules related to proliferation (36,37).
In conclusion, this study provides evidence that c-Fes is a novel downstream signaling molecule in FGF-2-regulated angiogenesis, affecting both chemotaxis and tube formation. Future studies will address the specific downstream signaling molecules activated by c-Fes in endothelial cells; identification of such molecules will provide a better understanding of the mechanisms underlying FGF-2-induced angiogenic responses.
Acknowledgments-We are grateful to Dr. Hiroshi Kikuchi at Daiichi   FIG. 7. Expression of wild-type or kinase-inactive c-Fes does not influence proliferation-related substrate phosphorylation or the kinetics of Ras/MAPK activation. A, tyrosine phosphorylation of FRS-2. Mock, K590E (clone 5-8), or WT (clone 6-1) c-Fes-transfected IBE cells were grown to confluence on 10-cm dishes and incubated in the presence or absence of 100 ng/ml FGF-2 for 8 min. FRS-2 protein was precipitated from clarified cell lysates with p13 suc1 -agarose beads and analyzed for tyrosine phosphorylation by anti-phosphotyrosine immunoblotting (IB). B, Shc tyrosine phosphorylation. Cells were treated as in A, and Shc proteins were precipitated (IP) with the anti-Shc polyclonal antibody and blotted with the anti-phosphotyrosine antibody. Blots were stripped and reprobed with the anti-Shc antibody to verify the presence of Shc protein in each lane. C, kinetics of MAPK activation by FGF-2. Cells were stimulated with 100 ng/ml FGF-2 for indicated periods and lysed with SDS-sample buffer. Lysates were blotted with anti-phospho-MAPK antibodies to determine the activation of MAPK. To examine the MAPK protein content in each sample, membranes were stripped and reprobed with anti-MAPK polyclonal antibodies. D, tyrosine phosphorylation of PLC-␥ by FGF-2 treatment. Cells were treated as in A, and PLC-␥ was immunoprecipitated from cell lysates and blotted with the anti-phosphotyrosine antibody. The blot stripped and reprobed with the anti-PLC-␥ antibody to verify the presence of PLC-␥ protein in each lane. Each experiment was repeated at least twice with comparable results.